How quickly could Iran go nuclear?

The New Yorker: Online Only: Content

The New Yorker, not known for being a right wing panic merchant, runs an interview this week with (journalist?) Steve Coll that is worth reading.

Mr Coll notes that on the question of how quickly Iran could have a nuke:

John Negroponte, the director of National Intelligence, has said, in his most recent public assessment, that the American intelligence community believes that Iran may acquire a nuclear capacity some time in the next decade, meaning from 2010 or 2011 onward. From my reporting, I gather that in private briefings the Bush Administration’s intelligence analysts focus on a five-to-seven-year window, although they emphasize that there’s a fair amount of uncertainty about this estimate. I think the one assertion that the intelligence community seems comfortable with is that it’s not this year or next year and probably not the year after that. However, the more that is discovered about Iran’s research, the more some analysts wonder whether Iran might be able to move faster than the official forecast indicates.

It gets worse, though:

[Interviewer] Once the centrifuges are working, how long will it take to make enough material for a bomb?

It depends on how many centrifuges you put into your plant. The math is fairly straightforward: a cascade of a hundred and sixty-four centrifuges can produce so many grams of highly enriched uranium in so much time if the centrifuges are operating around the clock. Iran has said that it intends to install three thousand of these centrifuges by the end of this year. That seems like an ambitious goal, but let’s assume the Iranians could achieve it. If they did, they could manufacture enough highly enriched uranium for a couple of bombs within a year if they operated those centrifuges around the clock. Most people don’t think they can pull that off, but that’s the scale of their operation at this point.